cursor-rules

A set of rules for keeping persistent memory across coding-agent sessions and sharing it across agents on the same project. It defines when to recall, save, and review project knowledge.

In plain words
What is it for?
Use it to initialize a session, recall relevant project memory, save important actions and decisions, record errors, and close a session with next steps.
Why use it?
It reduces repeated explanations and helps preserve decisions, lessons, and project context between sessions. It also requires recording errors and session summaries.

Cursor rule

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add rules/vbcherepanov/total-agent-memory/cursor-rules
Clone the repo
git clone --depth 1 https://github.com/vbcherepanov/total-agent-memory
Per session 499 This file is loaded in full into every session.
When invoked 499 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00499 $0.00499
Opus 5 $0.00249 $0.00249
Sonnet 5 $0.00100 $0.00100
Haiku 4.5 $0.00050 $0.00050

Measured yesterday against content hash f3d5b9b29194, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cursor-rules scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/memory-protocol/templates/cursor-rules.mdc · 59 lines

What it actually says

Memory Protocol (total-agent-memory v10.5)

You have access to a persistent cross-session memory via the memory MCP server. Knowledge survives between sessions and is shared across agents working on the same project.

The five non-negotiables

  1. Session start → session_init first, then memory_recall.
  2. Before any non-trivial task → memory_recall(query, project).
  3. After every significant action → memory_save immediately.
  4. On error / stuck → learn_error (or self_error_log).
  5. End of session → session_end with summary + next_steps + pitfalls.

Save discipline

ЧТО:        one-line summary
ПРОЕКТ:     project name
ФАЙЛЫ:      absolute paths to the key files
СТЕК:       language / framework / version
ПОДХОД:     3–7 key steps
НЮАНСЫ:     gotchas, edge cases

For decisions, add WHY (alternatives, trade-offs).

Tags must include ["reusable", "<tech>"] if the recipe applies elsewhere.

Recall discipline

  1. memory_recall(query, project=<current>) — current project first.
  2. Empty / contradicting → analogize(text=query, exclude_project=...).
  3. Still empty → context7 / WebSearch / first principles.
  4. Never guess a convention you can recall.

Self-improvement

  • 3 identical errors → STOP. memory_save(type='lesson', stuck=...) + ask the user with A/B/C options.
  • Reusable solution → memory_save(tags=['reusable', ...]).

Anti-patterns

  • Don't ask "should I save this?" — save without asking.
  • Don't batch saves until end of session — save now.
  • Don't recall and ignore — cite the id + 1-line gist before applying.
  • Don't save terminal dumps — distill to the digest above.
  • Don't guess a convention you can recall.

For the full tool reference, hooks documentation, and per-IDE setup see ~/total-agent-memory/skills/memory-protocol/references/.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 59 lines · 499 tokens per session scan A f3d5b9b29194

Subscribe to this mod's changes

cursor-rules is a cursor rule published in the GitHub repository vbcherepanov/total-agent-memory (66 stars, last pushed 4d ago), licensed MIT. It adds 499 tokens to every session, about $0.0025 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.